I04 NEW YORK STATE MUSEUM 



of 500 brachioles to some old specimens. With each added 

 brachiole there was also added a very short, rudimentary hydro- 

 spire consisting' at first of a membranous fold open externally. 

 As the hydrospires with their grooves are no farther apart in 

 adult specimens than in young ones [see pi. 5 fig. k-o of inner 

 surface of deltoids, where approximately 14 grooves lie side by 

 side in a width of 5 mm whether young or old plates are taken] 

 it follows that direct lateral growth was limited to the widening 

 of the plate by the addition of these new brachioles and their 

 hydrospires. 



Direct upward extension of the hydrospire and the portion of 

 the plate bearing it would soon bring the base of any one 

 brachiole up to the level of the former position of its next older 

 companion while the companion constantly maintained its former 

 superior relation by a similar upward extension of its own 

 hydrospire and plate portion. Thus the plate was indirectly 

 widened by upward growth of the portions consecutively added 

 at the lower angles. This upward extension in the early stages 

 of the development of the plate was no doubt as rapid as the 

 extension of the lower angles, but at a later stage the lower 

 angle extension was the more rapid. 



The rather remarkable parallelism of the hydrospires would 

 show that the basal plates of a young brachiole ceased to grow 

 as soon as a subsequent brachiole was added. That there was 

 a slight enlargement of the later formed brachioles is however 

 shown in plate 5, figure o, where the grooves of the hydrospire, 

 between 30 and 39, begin to show a very appreciable change in 

 direction. Another fragment not figured shows a still greater 

 extension of this angle of a plate and a change in the direction 

 of the grooves of more than 45 degrees. It is on this bit of evi- 

 dence that I have suggested that the 39 side grooves of plate 5, 

 figure o, may have been surpassed in this specimen by at least 

 II others. Its possessor must have shown a more remarkably 

 starfishlike form, when viewed axially than the specimen figured 

 in plate i. It must also have brought the ends of its ambulacra 

 more nearly down to a level with its base. 



The hydrospires and their plate portions were also extended 

 downward and the points of their origin became thus left along 

 a line still visible on the external surface of the plate and con- 

 necting the older central portion with the newer extension of 

 the angles. This line may be seen in plates i, 2 and 3, and is 

 also clearly shown in plate 5, figure j. This downward exten- 



